Copy__ THE UNIVERSITY OF MICHIGAN COLLEGE OF ENGINEERING DEPARTMENT OP ELICTRICAL ENGINEERING Radiation L1bor *Wry 4700-1-F = RL-2115 FINAL REPORT by Raymond F. Goodrich June 1962 Purchase Order 90044-680362 Contract With: International Business Machines Corporation Space Guidance Center Federal Systems Division Owego, New York Administered through: OFFICE OF RESEARCH ADMINISTRATION * ANN ARBOR

THE UNIVERSITY OF MICHIGAN 4700-1-F FINAL REPORT by Raymond F. Goodrich June 1962 Purchase Order 90044-680362 Prepared for International Business Machines Corporation Space Guidance Center Federal Systems Division Owego, New York w

THE UNIVERSITY OF MICHIGAN 4700-1-F I INTRODUCTION In the course of our discussions of the XB prototype radar antenna, two main problems arose: The first was the initial failure of the antenna to meet the pattern specifications due to the failure to take into account the inherent dispersive behavior of slot couplers. This problem was successfully solved by Dalmo-Victor. We offer some comments on this in section II below. The second problem was the "isodop" problem which imposes a restriction on the azimuthal beamwidth as long as the data processing does not make use of all the available information. We consider this in more detail in section II below. II ANTENNA ARRAY DESIGN The use of slot coupled elements in a linear array always implies a certain phase dispersion at each slot coupler. The dispersive behavior is an essential effect so that the amounts of dispersion at each slot need be parameters in designing for a given aperture illumination. In fact, over the given frequency range a knowledge of the dispersion and a then predictable change in "trombone" lengths has been used by Dalmo-Victor to obtain a constant dispersion over the frequency variation. The refinement of this scheme might lie in the use of a combination of corporate and trombone feeds. The point being that since the dispersive phase errors are cumulative on passing from slot to slot a reduction in the number of slots in series by the use of a parallel feed arrangement would reduce the cumulative phase error. We conclude our discussion of the antenna feed system by noting that the Dalmo-Victor system has made use of te slot dispersion as parameters un s mtthe design and the result has been most successful. 1

THE UNIVERSITY OF MICHIGAN 4700-1-F III THE ISODOP PROBLEM The isodop problem has been treated in some detail in a different context by Klauder, et. al. They were concerned with "chirp" radar systems and in some detail considered the problem of side lobes in the time domain. The problems are equivalent since chirp systems impose a linear frequency modulation on the signal while the high resolution radar schemes make use of the linear frequency modulation caused by the radial velocity of a target in the beam. Since the isodop problem is formally the same as the chirp problem we can then relax the beamwidth restriction as long as we can go to more sophisticated data processing as described by Klauder, et. al. That is to say, the aperture size is not a restriction in principle but the greater complexity of the processing can be a limitation. Hence for a given system restriction on the aperture, the processing can be specified; for a given processing scheme, the minimum aperture is fixed. * Bell System Technical Journal, 39, pp. 745-808 (1960). 2